HILLEL'S TECH CORNER: CytoReason: Where AI meets drug development

CytoReason’s technology uses a machine learning model to build a molecular level understanding of tissues in diseases on the cellular level.

CytoReason. (photo credit: Courtesy)
(photo credit: Courtesy)
Is it just me or do you also feel that we should have better tools and technology today to know what treatments will work on what patients and what will not? What I mean is, nothing moves in the world of pharmaceuticals without clinical trials and even the words “clinical trials” rub me the wrong way.
Why do we have to “try” things to know if they will work? Not to mention that when there is “trial,” that means it is accompanied by “error.” If you know someone who participated in a clinical trial only to discover, after months, that the treatment was totally ineffective for them and their condition, you surely know how devastating that can be.
Shouldn’t technology be able to predict how accurate certain treatments will be for specific patients, with at least some accuracy?
Even after the clinical trials are over, I still ask myself the question daily – why don’t we have a cure for so many conditions? We have self-driving cars, autonomous drones and human-like robots, yet somehow when it comes to many cancers, and new viruses such as COVID-19, we fall short. Why?
Well, the obvious answer is that clinical research and trials are complex and take between one to two decades to complete (in the best case scenario) before getting market approval. And if things fall apart midway? Back to square one. Not to mention that costs related to the failed trials still need to be paid. And as we all know, one size does not fit all when it comes to medical drugs on humans, and there is zero room for error in each individual case.
This is precisely the challenge CytoReason, a Tel Aviv based company, set out to overcome.
CytoReason was founded in 2016 by researchers and scientists from the Technion-Israel Institute of Technology. It was founded with the goal of leveraging new technology to address the increasing costs and complexity of bringing new therapies to market. CytoReason’s proprietary data is fully derived from human clinical trials, and coupled with their machine learning technology, it allows the company to build an unparalleled understanding of disease biology and simulate the effects of drug treatments, facilitating the discovery and development of more effective drugs. These simulations are applicable in a multitude of areas from cancer immunotherapy, and autoimmune, neurodegenerative and infectious disease research. CytoReason’s computational models of the human body are used today by the largest pharma companies for drug discovery and development.
In fact, in 2019, CytoReason formed a collaboration agreement with Pfizer Inc. (NYSE:PFE) that will leverage CytoReason’s cell-centered disease models to generate mechanistic understanding of disease, leading to novel insights. CytoReason has also collaborated with British multinational GlaxoSmithKline (gsk) and with the Parker Institute for Cancer Immunotherapy, a leading US research institution that coordinates cancer research efforts between scientists, clinicians and industry partners.
Based on more than 10 years of research, CytoReason’s technology uses a machine learning model to build a molecular level understanding of tissues in diseases on the cellular level. The company’s platform studies disease biology and how it is affected by drugs by leveraging proprietary molecular level clinical trial data it obtains from its pharmaceutical collaborators.
Their rapid success comes as no surprise, as the company’s leadership is highly skilled and experienced in pharma drug discovery, AI and machine learning technologies, with more than 70% of the company’s headcount holding PhDs in relevant fields of science.
David Harel, the CEO of CytoReason, spent the first part of his career in private equity, focusing on strategy and financing of growth companies in industrial and Healthcare IT markets. CytoReason’s CSO Shai Shen-Orr is also Head of the Systems Immunology and Precision Medicine Lab at the Technion School of Medicine. Shen-Orr’s research harnesses the revolution in computing power to integrate data, computational methodologies and machine learning approaches to create molecular-level models of disease, tissue and treatment environments. These models are capable of uncovering high-resolution insights into cellular and molecular associations, revealing novel targets, indications, combinations and biomarkers. CytoReason’s VP of engineering, Yuval Kalugny, is responsible for leading the CytoReason “Core Team” to solidify their position in computational, software and algorithmic development. He previously served as a captain in the elite 8200 Military Intelligence unit of the IDF.
CytoReason’s founders identified deficiencies of two essential factors needed to accelerate biological discovery for drug development. The first factor is to decipher complex human disease biology one needs a multitude of human molecular data, a lack of which exists even in the world’s largest pharma companies, which predominantly still rely on animal models for building an understanding of disease.
The second factor is that getting impactful biological insights about disease and drug development from this data requires integration of on-point algorithmic solutions with a strong understanding of biology, a combination hard to find, and therefore either performed sub-par or by a select few experts in a non-scalable manner. Moreover, most experts never have enough data at hand to extract the full extent of the biological discoveries waiting to be found.
The CytoReason platform learns with each new clinical trial it sees, continuously improving its accuracy of capturing disease biology. By learning from data across the pharmaceutical industry, the CytoReason models are in a unique position to understand how the different treatments affect the disease and garner novel insights critical for the drug development process, including robust target discovery, drug response biomarkers and selection of the appropriate disease indication for novel drugs.
To date, CytoReason’s technology has yielded five pending patents, 10 commercial and scientific collaborations, and 17 peer reviewed publications in top tier journals. Fully applicable to cancer immunotherapy, autoimmune, neurodegenerative and infectious disease research, CytoReason is at the cutting edge of society’s boldest attempts to improve health outcomes through better understanding of the immune system.
Between CytoReason’s incorporation of proprietary pharma clinical trial data; algorithms that turn data into mechanistic understanding of biology; and machine learning capabilities at scale, CytoReason gives biologists big data with the speed and accuracy that has long been needed by humankind.
We are all waiting for a breakthrough with so many conditions including, but not limited to cancer, COVID-19, and many others. When that breakthrough is achieved and CytoReason is at the forefront of it, I, for one, won’t be that surprised.